LAUSR

OBJECTIVE There is a paucity of good-quality evidence comparing direct surgical (DS) with endovascular/hybrid (EVH) revascularization for aortoiliac occlusive disease (AIOD). We aimed to perform a meta-analysis of studies comparing DS and EVH revascularization for AIOD. METHODS PubMed, Ovid MEDLINE, Cochrane, and Embase databases were searched for studies comparing DS and EVH revascularization for AIOD from 2000 to 2018. Risk of bias assessment was performed using the Methodological Index for Non-Randomized Studies. Demographics, clinical presentation, and comorbidities of the two groups were compared. Kaplan-Meier curves from selected studies were digitized with WebPlotDigitizer. Meta-analysis was conducted using Review Manager, and outcome measures were compared. Subgroup analysis was performed for primary patency in the EVH group. RESULTS Eleven observational studies were identified comparing a sample of 4030 patients. The median Methodological Index for Non-Randomized Studies score was 19 of 24. A total of 1679 and 2351 patients underwent DS and EVH techniques, respectively. No significant difference was found between means for sex, claudication, rest pain, tissue loss, preoperative ankle-brachial pressure index, and TransAtlantic Inter-Society Consensus C and D lesions in the two groups averaged across studies. However, the DS group had significantly younger patients (average age, 61.83 vs 66.77; P = .0011). The risk factors of the two groups, such as smoking, diabetes, ischemic heart disease, hypertension, hyperlipidemia, renal failure, and chronic lung disease, were comparable. Average hospital stay was significantly higher for the DS group (7.76 days vs 3.12 days; P = .025). Change in ankle-brachial pressure index, 30-day mortality, and 30-day graft/stent thrombosis were not significantly different for the groups. Overall, primary patency for a median follow-up of 50 months favored the DS group (hazard ratio [HR], 0.51; confidence interval [CI], 0.36-0.73; P = .0002). There was moderate heterogeneity among studies (I2 = 46%). The HR for the subgroup for which endovascular procedures were combined with common femoral endarterectomy was 0.43 compared with 0.88 for endovascular revascularization alone. Limb salvage was similar in both groups (HR, 1.10; CI, 0.74-1.64; P = .63), but overall survival after the procedure favored the DS group (HR, 0.75; CI, 0.60-0.94; P = .01; I2 = 0%). CONCLUSIONS Moderate-quality studies showed that DS revascularization had significantly better primary patency than EVH revascularization for AIOD, although DS patients were younger and may have differed on other confounding variables. Both techniques had similar limb salvage rates, and the primary patency was better for endovascular revascularization combined with common femoral endarterectomy than for endovascular revascularization alone.